ML19351D305
| ML19351D305 | |
| Person / Time | |
|---|---|
| Site: | 05000447 |
| Issue date: | 09/29/1980 |
| From: | Teh-Chiun Su Office of Nuclear Reactor Regulation |
| To: | Kniel K Office of Nuclear Reactor Regulation |
| Shared Package | |
| ML19351D306 | List: |
| References | |
| NUDOCS 8010090498 | |
| Download: ML19351D305 (4) | |
Text
'
[
- ', 't,
. UNITED STATES as NUCLEAR REGULATORY COMMISSION
[ ;.! )A,1 }
WASHING TON, D. C. 20555
-[s[
SEP Z 3 1980 Task' Action Plan A-39 Docket No. 50-447 MEMORANDUM FOR:
Karl Kntel, Chief Generic Issues Branch Division of Safety Technology, NRR FROM:
T. M. Su, A-39 Task Manager Generic Issues Branch Division of Safety Technology, NRR APPLICANT:
General Electric Company for GESSAR-238
SUBJECT:
MEETING llITH REPRESENTATIVES OF GENERAL ELECTRIC TO DISCUSS MULTIPLE-SRV'S LOADS Backcround In early 1978, the General Electric Company proposed a methodology to determine the multiple safety / relief valves (SRV) loads used for NSSS equipment evaluation. This method uses Monte Carlo techniques to calculate the phase relationship of the individual air bubbles formed in the suppression pool during multiple SRV discharge events. The random variables that are utilized in the method include (1) SRV Setpoint Tolerance (SPT), (2) Valve Openin Reactor Vessel Pressure Rise Rate (PRR), and (4) g Time (VOT), (3)
Quencher Bubble Frequency (QBF). The staff and oJr Consultants have had discussions with the representatives of GE in caveral meetings since then. The purpose of this meeting was to disuss the questions that were raised by the staff and our consultants as a result of reviewing the information submitted by GE. These questions were transmitted to GE by letter dated July 22, 1980 from J. R. Miller (NRC) to G. Sherwood (GE).
A list of attendees and a copy of the meeting handouts are enclosed.
1 Sunnary The following surxiarizes the significant areas of the discussion:
1.
Mr. Steve Hucik of General Electric presented the results of analyses to demonstrate the conservatism of the safety / relief valve actuation (SRVA) methodology. The analysis was performed by using i
i 8 010 0 00 N-
k Karl Kniel SEP 2 s 1980 the Caorso test conditions for the four valve test series as input parameters for the SRVA method. Comparisons were then made between predicted values versus measured values in the form of Amplified Response Spectrum (ARS) (see Figure Pl. 1 of Mr. S. Hucik's handout).
The results show substantial margin over a wide range of bubble frequencies (2 to 200 Hz). GE indicated that the predicted high SRV pressure magnitudes are the primary contributor to this substantial conservatism. Since the ARS represents acceleration loads on equipment and piping, GE concluded that the SRVA provides conservative input for equipment design.
2.
GE noted that the method used for predicting SRV pressure magnitude in the comparison indicated above was based on the new methodology provided in the GESSAR docket. This new method was submitted for the staff review in the second quarter of 1980. The SRV loads calculated by this method will result in a load reduction of about 35 percent from that predicted by the current accepted methodology.
The staff indicated that review of this new method is currently scheduled to be completed by April 1981.
3.
GE also performed a sensitivity study on the random variables (STP, V0T, PRR and QBF). Results of this study showed that the forcing function is insensitive to the varied range of interest.
In particular, the variation in mean bubble frequency over the range from 7.5 to 10.1 HZ (-4 ; + 6.7d caused little change in the peak spectral 7
values and the envelope of the peaks.
4.
Dr. P. Valardani of GE presented the justification for selection of influencing parameters on bubble frequency. The parameters include air volume, pool temperature, bubble pressure, submergence and free surface. Some of these parameters are interrelated. How?ver, GE concluded that the most significant parameter is the air tolume initially stored in the SRV line. This conclusion was based on both full scale tests and half scale tests which were conducted over a wide range of all-parameters involved. Therefore, GE believes that the methodology is justified by using air volume as the sole parameter to vary bubble frequency.
5.
The staff and our consultants expressed concern on the significant variation of test data obtained from inplant tests in two BWR plants. This variation is clearly exhibited on the regression analysis (see Table 3.2.3 of GE response to Question 2(c), which is included in the attached handout). GE indicated that the exact reasons for the data variability are unknown because these inplant tests were uncontrolled tests. The source of data variability could be a result'of unknown air and steam composition in the line, valve leakage, and unknown water leg submergence.
A SEP 2 91980 Karl.Kniel-NRC Staff Coments T. Su of the NRC staff summarized the position of the staff and its consultants' on the SRVA methodology as follows:
1.
Based on the comparison between the predicted values versus the Caorso measured values, it is concluded that the SRVA methodology is conservative and acceptable. However, the final approval of the method is dependent upon the staff review on the infonnation which will be submitted by GE through the GESSAR docket.
2.
Since the comparison was based on the Caorso tests, which is a Mark II containment, we will require inplant tests for the Mark III containment to verify the validity of the bubble frequency range, bubble pressure magnitude and pressure attenuation used in the SRVA methodology.
GE indicated that the information will be submitted to NRC by early October,1980. With respect to inplant tests, GE also indicated that Taipower will perform SRV inplant tests on the first Mark III containment in the world by early spring 1981.
Y
~
T. M. Su A-39 Task Manager Generic Issues Branch Division of Safety Technology
Attachment:
~ Attendance List
ic, Attendance GE/NRC Meeting Mark III Multiple SRV Time Phasing September 10, 1980 Name Oraanization Larry Steinert GE T. M. Su NRC/ DST /GIB C-Economus BNL P. Huber MIT/BNL S. Finch-SUNY SB/BNL D. Freeman Decision Dynamics /GE A. Kanters GE/ STRIDE D. Ashley Power Technical Associates P. Stancavage GE/ Containment Forrest Hatch GE/ Mark III Containment T. Trocki GE/ Cont. Prog's.
Fred Reuter GE/ Mark III Containment Steve Hucik GE/ Containment Engr.
Joe Henry GE/ Containment Engr.
P. V. Valandani GE/SRV R. Post SJSU/ Consultant to GE D. F. Guvot Black & Veatch D. Braden GE/ Mark III Containment D. E. Hughes GE/ Int. Projects N. G. Gunther GE/SRV/ Containment Engr.
R. L. Beck BECHTEL Power Corporation M. R. Knight Mississippi P&L M. J. Anciaux GE y
a
-g--
-n.
y.w-
-